Effect of added polymer on the desiccation and healing of a geosynthetic clay liner subject to thermal gradients
Introduction
Geosynthetic Clay Liners (GCLs) have been extensively used in geotechnical and geoenvironmental engineering practice as barriers due to their extremely low hydraulic conductivities (<10−10 m/s) (Shackelford et al., 2000; Bouazza, 2002; Rowe, 1998, 2005). The key component in GCLs is the thin layer of bentonite clay, usually sandwiched between two geotextiles, which keeps the liner's hydraulic conductivity low when well hydrated by clean water (i.e., minimal total dissolved solids). However, such ideal conditions do not always occur in the field since GCLs are often not fully hydrated with clean water (Shackelford et al., 2000; Jo et al. 2001, 2005; Kolstad et al., 2004; Acikel et al., 2018; Gates et al., 2018; Kul and Ören, 2018; AbdelRazek and Rowe, 2019; Rowe, 2020) before being exposed to chemical hazards, wet-dry circles, freeze-thaw circles, high temperatures, and thermal cycles (Kraus et al., 1997; Lin and Benson, 2000; Bouazza, 2002; Rowe, 2005; Lu et al., 2018). In particular, containment facilities that expose GCLs to both high temperatures and aggressive organic/inorganic chemicals (e.g., landfills, solar ponds and brine ponds) present a challenge (Ghavam-Nasiri et al., 2019; Rowe, 2020; Yu and El-Zein, 2019)
According to Southen and Rowe (2005), high temperatures on top of GCLs result in a thermal gradient between the warm heat source on top and the cooler subsoil at the depth, which eventually drives the moisture down and can lead to dehydration of the GCL. This phenomenon was noticed in landfill barrier systems, where temperatures in the waste may be higher than 50–60 °C due to the biodegradation process (Rowe, 2005). Laboratory and field investigations further revealed that GCL in the bottom liner of landfills was at risk of desiccation, which increases with higher temperatures and lower subsoil water content (Southen and Rowe, 2005). Similar observations were reported for GCLs in double composite lining system (Azad et al., 2011). Investigations also indicated that higher risks of desiccation pertain to GCLs with lower initial water content prior to exposure to thermal gradients (Azad et al., 2011; Sarabadani and Rayhani, 2014). In facilities such as brine ponds and solar ponds where temperatures up to 90 °C have been reported during daytime (El-Sebaii et al., 2011; Bouazza et al., 2014), the smaller overburden stresses on the liner system, compared to landfill systems, means higher risks of desiccation (Southen and Rowe, 2005; Rowe and Verge, 2013; Hoor and Rowe, 2013). Further investigations of the GCLs in brine ponds confirmed that severe desiccations may occur regardless the types or manufacture qualities of GCLs (Ghavam-Nasiri et al., 2017; Ghavam-Nasiri, 2017; Yu and El-Zein, 2019). Results reported by Bouazza et al. (2017) suggested that by introducing air gaps as thermal barrier, between the source of heat and the underlying GCL (e.g., a geocomposite drain between a primary GMB liner and secondary composite liner), desiccation of bentonite in the GCL in the secondary liner may be reduced or even avoided. However, in laboratory experiments, Yu and El-Zein (2019) found that, although the introduction of an airgap reduced the temperature on top of the GCL from 78 °C to 45 °C, desiccation still occurred when the GCL was placed on highly permeable subsoils.
GCL that have dehydrated leaving desiccated bentonite would fail to function adequately if they cannot effectively heal upon rehydration. This is of particular concern if they are exposed to chemical solutions with high salinities. Although polymer treatment of Na-bentonite components in GCLs has been found to improve their chemical resistance (e.g., the polyacrylate modified bentonite reported by Scalia, 2012, and Scalia et al., 2011, 2013), little is known about how it affects the thermal desiccation behavior and healing performance of polymer-enhanced GCLs in this environment. This raises the question as to whether desiccation cracks formed in GCLs exposed to thermal gradients can self-heal upon rehydration. The literature on expansive clay suggests that self-healing may be affected by exposure to high temperatures. For example, Romero et al. (2003) found that when temperature increased from 22 to 80 °C, there was a significant decrease in swelling pressure in expansive clays. Estabragh et al. (2016) reported that the swelling of bentonite under rewetting decreased when it was previously exposed to higher temperatures (>50 °C).
On the other hand, hydraulic conductivity tests using distilled water on desiccated GCLs have shown that the self-healing capacity and hydraulic performance of the bentonite were not significantly influenced by their thermal history (Southen and Rowe, 2005; Ghavam-Nasiri, 2017; Yu and El-Zein, 2019). However, while useful as far as it goes, this really does not address the question of self-healing when thermally desiccated GCLs are permeated by a more aggressive fluid because the effects of chemicals on GCLs were not included in the three studies mentioned above. When bentonite clays are exposed to ionic solutions, cation exchange occurs and the diffuse double layer (DDL) on clay mineral surfaces becomes thinner, leading to more free-flow channels and higher permeability (Mesri and Olson, 1971; Mitchell and Soga, 2005). Previous studies also showed that hydraulic conductivities of GCLs increased in the presence of salt in the porewater (Petrov and Rowe, 1997; Shackelford et al., 2000; Jo et al., 2005; Benson and Meer, 2009). Therefore, the extent to which self-healing is effective when the thermally desiccated GCL is hydrated with chemically aggressive solutions such as brine, remains unknown.
Meanwhile, efforts have been put into modifying bentonite clays to enhance their chemical compatibility in a cation-rich environment, specifically in order to maintain their swelling potential and low permeabilities (Onikata et al., 1999; Katsumi et al., 2008; Malusis and McKeehan, 2013; Scalia, 2012; Scalia et al., 2011, 2013; De Camillis et al., 2016, 2017 and 2018). It was reported that GCLs containing polymer-enhanced bentonite exhibited better chemical compatibility as assessed by hydraulic conductivity tests (Scalia, 2012; Scalia et al., 2011, 2013; Athanassopoulos et al., 2015; Arndt et al., 2015). However, nothing is known about the desiccation and healing behaviour of polymer enhanced bentonite in GCLs.
The objective of this study is to assess the desiccation and healing potential of GCLs, with and without polymer modification, when exposed to thermal gradients and then permeated with distilled or highly saline water. Two types of GCL were subjected to thermal gradients in a series of instrumented column tests that replicated conditions similar to those found in brine ponds, namely, low overburden pressure, high thermal gradients, and subsequent permeation with a salt solution (brine). One GCL was made of Na-bentonite modified with polyacrylamide, while the other contained only the original Na-bentonite without polymer modification. The two GCLs were otherwise identical. The influence of temperature on the desiccation of the two types of GCLs was investigated by monitoring water loss and desiccation cracking. Some desiccated specimens were then subjected to direct hydraulic conductivity tests with distilled water while others were permeated with a synthetic brine.
Section snippets
GCLs
Two needle-punched, thermally treated GCLs with fine granular Na bentonite, denoted as GCL_A and GCL_B, were studied. Both GCLs had a nonwoven cover geotextile and scrim reinforced nonwoven carrier geotextile and a similar value (4200–4300 g/m2) for the air-dry bentonite mass per unit area (Table 1). The only notable difference between the two GCLs was that the granular Na-bentonite of GCL_B had been modified with a polyacrylamide polymer.
The water retention characteristics of the two GCLs were
Temperature effects on dehydration of GCLs
The dehydration processes of GCL_A and B were investigated at temperatures directly above the GMB ranging from 35 to 60 °C using the medium-scale columns. Measurements of specimens’ weights during the tests allowed moisture variations of GCL samples to be calculated as shown in Fig. 8. From Fig. 8b it can be seen that apart from the GCL_A and GCL_B specimens subjected to 35 °C, which had lower initial water contents (~87%), all other specimens had very close water contents (96.5% ≤ ω ≤ 100.7%)
Conclusions
A series of soil column laboratory tests were conducted in which GCLs containing standard and polymer-enhanced Na-bentonite were allowed to hydrate to 80–100% gravimetric and then subjected to a thermal gradient for two weeks so simulate field conditions of a liner in a brine pond. The GCLs were removed from the columns and the hydraulic conductivity measured using, either distilled water or 0.325 mol/L Na+ synthetic brine (with NaCl, Na2CO3 and NaHCO3) found in brine ponds related to shale-gas
Acknowledgments
This work was supported in part by the Australian Research Council through the Discovery Projects Scheme (DP170104192). We would like to thank Professor David Airey from University of Sydney for providing advice on some of the experiments. We would like to thank Dr. Francois Guillard from University of Sydney for providing help on x-ray imaging tests on GCLs and Terrafix Geosynthetic Inc. Canada for providing the GCLs tested. We are also grateful for the support provided by technical staff at
References (67)
- et al.
Interface transmissivity of conventional and multicomponent GCLs for three permeants
Geotext. Geomembranes
(2019) - et al.
Insufficient initial hydration of GCLs from some subgrades: factors and causes
Geotext. Geomembranes
(2018) - et al.
Laboratory investigation of thermally induced desiccation of GCLs in double composite liner systems
Geotext. Geomembranes
(2011) Geosynthetic clay liners
Geotext. Geomembranes
(2002)- et al.
Heat and moisture migration in a geomembrane–GCL composite liner subjected to high temperatures and low vertical stresses
Geotext. Geomembranes
(2014) - et al.
Heat mitigation in geosynthetic composite liners exposed to elevated temperatures
Geotext. Geomembranes
(2017) - et al.
Hydraulic conductivity and swelling ability of a polymer modified bentonite subjected to wet–dry cycles in seawater
Geotext. Geomembranes
(2016) - et al.
Effect of wet-dry cycles on polymer treated bentonite in seawater: swelling ability, hydraulic conductivity and crack analysis
Appl. Clay Sci.
(2017) - et al.
Forward osmosis as a pre-treatment for treating coal seam gas associated water: flux and fouling behaviours
Desalination
(2017) - et al.
Membrane distillation and membrane electrolysis of coal seam gas reverse osmosis brine for clean water extraction and NaOH production
Desalination
(2016)